Protein misfolding and aberrant self-assembly into toxic species ranging from small oligomers to large amyloid fibrils are pathologically linked to several neurodegenerative disorders such as Alzheimer?s disease. Conformational antibodies with specificity for protein aggregates are important for investigating the role of different types of aggregates in neurodegenerative diseases as well as for potentially treating these debilitating diseases. It has, however, been extremely difficult to generate conformational antibodies against protein aggregates due to several factors: i) the limitations of immunization (lack of control over antigen presentation); ii) the fixed number of binding sites per antibody, which limits the use of polyvalency for targeting multimeric protein aggregates; and iii) the difficulty in using nave antibody libraries to obtain conformational antibodies via in vitro selection methods. It is particularly challenging to generate conformational antibodies specific for tau protein aggregates (associated with Alzheimer?s disease) due to the large size of tau (up to 441 residues) and its multiple isoforms. To address these challenges, we have recently developed a rational, systematic and broadly applicable approach for generating domain antibodies with specificity for protein aggregates. The goal of this proposal is to use this approach to generate conformational domain antibodies specific for tau oligomers and fibrils, and to use these antibodies to evaluate the relative importance of different types of tau aggregates in mediating tau pathology in animal models. Our proposed approach builds on our collective experience in: i) designing domain antibodies with specificity for protein aggregates based on homotypic interactions between identical peptide motifs; ii) enhancing the affinity and specificity of domain antibodies using directed evolution; iii) designing polyvalent molecules that bind to oligomeric proteins with high affinity and specificity; iv) assembling and isolating tau oligomers and fibrils; and v) evaluating the ability of antibodies to prevent and reverse pathology in tau animal models. In Aim 1, we will test our hypothesis that domain antibodies with enhanced conformational specificity and affinity for tau oligomers and fibrils can be readily selected from antibody libraries with tau amyloidogenic peptides grafted into the main binding loop (CDR3). Next, in Aim 2, we will evaluate our hypothesis that polyvalency can be used to increase the conformational specificity and affinity of tau domain antibodies selected in Aim 1 by generating polyvalent versions in a manner that affords control over the number and spacing of domain antibodies. Finally, in Aim 3, we will test the ability of the most specific and inhibitory tau polyvalent domain antibodies generated in Aim 2 to inhibit tau seeding, spontaneous aggregation and pathology in vivo using tau mouse models. Significant outcomes of our studies will be systematic methods for generating polyvalent domain antibodies specific for different types of protein aggregates and evaluation of the relative importance of tau oligomers and fibrils in mediating pathology, which is important for understanding and potentially treating Alzheimer?s disease and other tauopathies.